Panel Discussion on Industrial Policy

1
Panel Discussion on Industrial Policy

• Ann Harrison
• Trade and International Integration
• Development Economics Research Group
• September 14, 2009

2
Introduction

• My discussion today is based on a forthcoming
  article co-authored with Andres Rodriguez-Clare
  for the next Handbook of Development Economics
• We were asked by Dani Rodrik and Mark Rosenzweig
  to answer the following question

3
When Does Industrial Policy Work?

• .Growth was not a passive, trickle-down strategy
  for helping the poor. It was an active, pull-up
  strategy instead. It required a government that
  would energetically take steps to accelerate
  growth, through a variety of policies including
  building infrastructure such as roads and ports
  and attracting foreign funds.
• Jagdish Bhagwati, In Defense of Globalization
  (2004)

4
Outline

• What do we mean by industrial policy (IP)?
• Theoretical justification for IP in trade
• The evidence on infant-industry protection
• Cross-country evidence on tariffs, trade, and
  growth
• IP and Foreign Direct Investment
• Policy Recommendations soft versus hard IP

5
What do we mean by Industrial Policy ?

• Working definition any intervention which shifts
  incentives away from policy neutrality
• A broad conception of IP, which spans a range of
  policies
• Tariffs
• Tax breaks
• Trade promotion

6
Distinguishing between Hard and Soft IP

• Hard Industrial Policy
• Tariffs
• Subsidies to specific sectors
• Tax breaks for foreign investors
• Domestic content requirements

• Soft Industrial Policy
• Special Economic Zones offering lower cost
  infrastructure
• Roads and ports designed to increase trade
• Special Credit for exporters (Trade Credit)
• Promoting clusters in order to export

7
When does IP make sense in theory?

• Whenever there are industry-level (Marshallian)
  externalities, and a sector has a latent
  comparative advantage or there are excess profits
• Sector receiving support must eventually be able
  to compete on international markets
• When the (discounted future) benefits from
  intervention exceed the costs of the distortion
• Not enough to show that favored sector grew
  faster
• Need to be able to show that welfare is higher
  (but no one does the welfare calculations in
  general)

8
What do we mean by a latent comparative
advantage?
9
How to resolve this debate? Answer lies in the
facts

• Evidence on infant industry protection
• Cross-country evidence on Tariffs, Trade Shares,
  and Growth
• Foreign investment and IP
• Biggest limitation of existing research no
  evidence that intervention in trade for IP
  reasons even exists

10
Empirical Evidence on infant-industry protection

• Evidence is mixed suggests that getting
  interventions right is difficult
• Industry-specific case studies
• Yes, welfare increased due to intervention
• Aircraft in Europe (Baldwin and Krugman (1989))
• Steel rail industry in the USA (Head (1994))
• Production of electricity from wind power (Hansen
  et al (2003))
• No, welfare fell as a consequence of
  intervention
• Semi-conductors in Japan (Baldwin and Krugman
  (1986))
• Tinplate in the United States ( Irwin (2000))
• Computers in Brazil (Luzio and Greenstein (1995))
• Cross-industry studies
• Removal of protection often generates within-firm
  and within-industry productivity gains
• But studies typically fail to measure the impact
  of policies, focusing instead on outcomes (like
  trade shares).

11
Cross-country studies

• Strong positive relationship between trade shares
  and growth
• Weak, generally insignificant relationship
  between tariffs and growth on average
• If IP works, expect a positive relationship
• If IP doesnt work, expect a negative,
  significant effect
• So evidence in aggregate is not conclusive for
  tariffs
• However, the PATTERN of protection matters
• tariffs on investment goods negatively and
  significantly affect growth
• WHAT you export matters

12
Taken from Easterly, Reshef, and Schwenkenberg
(2009)
13
So what to conclude from this mixed evidence?

• We know externalities exist (Rosenthal (2004))
  but exploiting them through hard IP is not
  easy
• If agglomeration economies are important, then IP
  in small domestic markets is likely to fail
• Any interventions associated with increasing the
  share of trade in GDP are more likely to succeed
• In practice, this means that
• Intervention should be oriented towards
• Sectors with a latent comparative advantage (Lin
  (2009))
• Sectors with large externalities or coordination
  failures
• IP should be in the form of export promotion
  (Trade Facilitation, Aid for Trade, Trade Credit)
  instead of import protection

14
Policy Implications for IP and Trade

• Who is doing IP matters (need strong institutions
  to prevent capture)
• What is being promoted makes a big difference
• Sectors with strong externalities (learning by
  doing spillovers to other sectors)
• Sectors with a latent comparative advantage
• When to promote emerging not declining sectors
• How to promote soft, not hard IP Instead of
  tariffs, we propose programs and grants to help
  particular clusters by increasing the supply of
  skilled workers, encouraging technology adoption,
  and improving regulation and infrastructure.
• Warning Agglomeration may be necessary but not
  sufficient for increased productivity.
  Subsidizing the software sector may not generate
  a Silicon Valley.

15
Foreign Investment and Industrial Policy

• All countries promote incoming foreign
  investment, but this is rarely referred to as IP.
• In 1998, 103 countries offered tax concessions to
  foreign companies setting up facilities within
  their borders.
• Countries frequently offer other benefits, such
  as free or subsidized infrastructure.
• Foreign investment policies as examples of
  successful IP (back to what, when, and how)

16
Many countries promote FDI selectively (From
Alfaro and Charlton)
17
Specific Suggestions for Soft IP (1)

• Regulations to enforce higher quality standards
• Public investment in specific infrastructure
  projects when there are large investment
  complementarities
• Attracting FDI through provision of
  infrastructure
• Scholarships for studies abroad in areas
  important for diversifying clusters but with thin
  markets
• Technical assistance, prizes, grants for projects
  proposed by organized producers and performed by
  local research centers

18
Specific Suggestions for Soft IP (2)

• Dont expect governments to identify coordination
  failures, but invite sector and cluster
  organizations to come forward
• If such organizations are weak, provide support
  to sectors that want to initiate or improve their
  organizations
• In general, avoid price interventions to
  reallocate resources but use existing clusters to
  identify effective interventions.
• Public-private collaboration is crucial

19
Conclusions

• Broader vision of IP
• Interventions must be consistent with growing
  trade
• Economists generally favor IP for foreign
  investors some evidence that such IP may work
  because it targets new or export oriented,
  growing sectors with externalities
• Support for soft IP rather than hard IP
• All countries engage in IP question is not
  whether to do IP, but how